1. Field of the Invention
The present invention relates to an image processing apparatus that switches processing methods in accordance with the attribute of the image, and more particularly, to an image processing apparatus that can determine dot areas with superior accuracy.
2. Description of the Related Art
In a digital copying machine that reads an image using a CCD sensor and breaks it down to pixels to perform image processing, because the optimal image processing method varies depending on the nature of the image, i.e., whether the image comprises letters, a photo or dots, different image processing methods are used based on the result of a determination regarding the attribute of the image. When a dot image is copied, moire may occur due to the interference between the resolution of the CCD sensor (i.e., the sampling frequency) and the dot frequency of the original document. When moire occurs, the image quality deteriorates. Consequently, the dot areas existing in the original document should be accurately determined, and image processing that will reduce the generation of moire should be performed.
To that end, in an image processing apparatus incorporated in a digital copying machine or similar apparatus, dot area determination has conventionally been carried out. In one example of such determination, isolated points (i.e., largest-value points and smallest-value points), which are the characteristic of a dot area, are focused on, the number of isolated points existing in a prescribed area are counted, and determination of a dot area is made based on the counted number of isolated points. In another example, the distances between isolated points are calculated, and when the calculated distances exhibit periodicity, a dot area is determined to exist.
However, the conventional image processing apparatus described above has the problem that, if the resolution increases, dot area determination is not carried out accurately. This is caused because the accuracy of dot area determination by a conventional image processing apparatus depends on the accuracy of the detection of isolated points. As the resolution of the image increases, it becomes increasingly difficult to detect isolated points in dot images having a small number of lines (i.e., dot images having 100 L (lines per inch) or 85 L, for example). Where the resolution increases, the size of one dot (pixel) in the detection filter by which to detect isolated points becomes small relative to the dots. As a result, the difference in density from the surrounding pixels becomes small (in some cases, the difference in density completely disappears), making it difficult to determine isolated points. Consequently, when the resolution increases, the accuracy of dot area determination falls.
In addition, there are situations in which image noise and hand-written letter images are erroneously determined to belong to a dot area. This is caused by the tendency of image noise and hand-written letter images to have a density difference of a minimum prescribed value from their surrounding pixels, like isolated points.